[PATCH] media: rcar-csi2: Update D-PHY startup on V4M

[PATCH] media: rcar-csi2: Update D-PHY startup on V4M

[Niklas Söderlund]

The latest datasheet (Rev.0.70) updates the D-PHY start-up sequence for
D-PHY operation. Unfortunately the datasheet do not add any additional
documentation on the magic values.

This have been tested together with the MAX96724 available on the single
board test platform and it works as expected.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
---
 drivers/media/platform/renesas/rcar-csi2.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/media/platform/renesas/rcar-csi2.c b/drivers/media/platform/renesas/rcar-csi2.c
index 27ffdd28cbf7..5f62f26a6b6f 100644
--- a/drivers/media/platform/renesas/rcar-csi2.c
+++ b/drivers/media/platform/renesas/rcar-csi2.c
@@ -1349,15 +1349,15 @@ static int rcsi2_init_common_v4m(struct rcar_csi2 *priv, unsigned int mbps)
 	static const struct phtw_value step2[] = {
 		{ .data = 0x00, .code = 0x00 },
 		{ .data = 0x80, .code = 0xe0 },
-		{ .data = 0x01, .code = 0xe1 },
+		{ .data = 0x31, .code = 0xe1 },
 		{ .data = 0x06, .code = 0x00 },
-		{ .data = 0x0f, .code = 0x11 },
+		{ .data = 0x11, .code = 0x11 },
 		{ .data = 0x08, .code = 0x00 },
-		{ .data = 0x0f, .code = 0x11 },
+		{ .data = 0x11, .code = 0x11 },
 		{ .data = 0x0a, .code = 0x00 },
-		{ .data = 0x0f, .code = 0x11 },
+		{ .data = 0x11, .code = 0x11 },
 		{ .data = 0x0c, .code = 0x00 },
-		{ .data = 0x0f, .code = 0x11 },
+		{ .data = 0x11, .code = 0x11 },
 		{ .data = 0x01, .code = 0x00 },
 		{ .data = 0x31, .code = 0xaa },
 		{ .data = 0x05, .code = 0x00 },
@@ -1370,6 +1370,11 @@ static int rcsi2_init_common_v4m(struct rcar_csi2 *priv, unsigned int mbps)
 		{ .data = 0x05, .code = 0x09 },
 	};
 
+	static const struct phtw_value step3[] = {
+		{ .data = 0x01, .code = 0x00 },
+		{ .data = 0x06, .code = 0xab },
+	};
+
 	if (priv->info->hsfreqrange) {
 		ret = rcsi2_set_phypll(priv, mbps);
 		if (ret)
@@ -1400,7 +1405,7 @@ static int rcsi2_init_common_v4m(struct rcar_csi2 *priv, unsigned int mbps)
 			return ret;
 	}
 
-	return ret;
+	return rcsi2_phtw_write_array(priv, step3, ARRAY_SIZE(step3));
 }
 
 static int rcsi2_start_receiver_v4m(struct rcar_csi2 *priv,
-- 
2.47.1

Re: [PATCH] media: rcar-csi2: Update D-PHY startup on V4M

[Sakari Ailus]

Hej Niklas,

On Tue, Dec 10, 2024 at 04:54:00PM +0100, Niklas Söderlund wrote:
> The latest datasheet (Rev.0.70) updates the D-PHY start-up sequence for
> D-PHY operation. Unfortunately the datasheet do not add any additional
> documentation on the magic values.
> 
> This have been tested together with the MAX96724 available on the single
> board test platform and it works as expected.
> 
> Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>

This looks like a fix. Is there a need to backport it?

-- 
Regards,

Sakari Ailus

Re: [PATCH] media: rcar-csi2: Update D-PHY startup on V4M

[Niklas Söderlund]

Hej Sakari,

On 2024-12-12 07:20:15 +0000, Sakari Ailus wrote:
> Hej Niklas,
> 
> On Tue, Dec 10, 2024 at 04:54:00PM +0100, Niklas Söderlund wrote:
> > The latest datasheet (Rev.0.70) updates the D-PHY start-up sequence for
> > D-PHY operation. Unfortunately the datasheet do not add any additional
> > documentation on the magic values.
> > 
> > This have been tested together with the MAX96724 available on the single
> > board test platform and it works as expected.
> > 
> > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
> 
> This looks like a fix. Is there a need to backport it?

It's an update to the start-up sequence. Both the old and new sequence 
works in testing. As the datasheet don't document the magic values it's 
hard to tell if this is a real fix or not.

On the other hand, the DTS changes needed to use this device on R-Car 
V4M are still pending upstream merge so we have no in-tree users of this 
device. I don't think we need to backport this change.

-- 
Kind Regards,
Niklas Söderlund

Re: [PATCH] media: rcar-csi2: Update D-PHY startup on V4M

[Sakari Ailus]

Hejssan,

On Thu, Dec 12, 2024 at 11:04:45AM +0100, Niklas Söderlund wrote:
> Hej Sakari,
> 
> On 2024-12-12 07:20:15 +0000, Sakari Ailus wrote:
> > Hej Niklas,
> > 
> > On Tue, Dec 10, 2024 at 04:54:00PM +0100, Niklas Söderlund wrote:
> > > The latest datasheet (Rev.0.70) updates the D-PHY start-up sequence for
> > > D-PHY operation. Unfortunately the datasheet do not add any additional
> > > documentation on the magic values.
> > > 
> > > This have been tested together with the MAX96724 available on the single
> > > board test platform and it works as expected.
> > > 
> > > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
> > 
> > This looks like a fix. Is there a need to backport it?
> 
> It's an update to the start-up sequence. Both the old and new sequence 
> works in testing. As the datasheet don't document the magic values it's 
> hard to tell if this is a real fix or not.
> 
> On the other hand, the DTS changes needed to use this device on R-Car 
> V4M are still pending upstream merge so we have no in-tree users of this 
> device. I don't think we need to backport this change.

Ack. Thanks for confirming this.

-- 
Sakari Ailus